zaclys/searxng
1
0
Fork 1
mirror of https://github.com/searxng/searxng synced 2024-01-01 19:24:07 +01:00
Code Issues Projects Releases Packages Wiki Activity

[mod] mudularize & document searx.results

Browse source 
The intention of this patch is to improve modularization & documentation of the
implementations about the *result* items.

  This patch does not contain any functional change!

Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Markus Heiser 2022-06-21 13:14:11 +02:00
parent cee586029c
commit 3d473a773d
15 changed files with 926 additions and 256 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

116
docs/dev/engine_overview.rst
View file

@ -120,6 +120,11 @@ module:
Making a Request
================
.. sidebar:: info
- Demo of :py:obj:`request(query, params)
<searx.engines.demo_online.request>` function.
To perform a search an URL have to be specified. In addition to specifying an
URL, arguments can be passed to the query.
@ -205,109 +210,18 @@ following parameters can be used to specify a search request:
raise_for_httperror bool True by default: raise an exception if the HTTP code of response is >= 300
=================== =========== ==========================================================================
.. _engine response:
.. _engine results:
.. _engine media types:
Making a Response
=================
Media Types
===========
.. sidebar:: info
Each result item of an engine can be of different media-types. Currently the
following media-types are supported. To set another media-type as ``default``,
the parameter ``template`` must be set to the desired type.
- Demo of a :py:obj:`response(resp) <searx.engines.demo_online.response>` function.
.. table:: Parameter of the **default** media type:
:width: 100%
In the ``response`` function of the engine, the HTTP response (``resp``) is
parsed and a list of results is returned.
========================= =====================================================
result-parameter information
========================= =====================================================
url string, url of the result
title string, title of the result
content string, general result-text
publishedDate :py:class:`datetime.datetime`, time of publish
========================= =====================================================
.. table:: Parameter of the **images** media type:
:width: 100%
========================= =====================================================
result-parameter information
------------------------- -----------------------------------------------------
template is set to ``images.html``
========================= =====================================================
url string, url to the result site
title string, title of the result *(partly implemented)*
content *(partly implemented)*
publishedDate :py:class:`datetime.datetime`,
time of publish *(partly implemented)*
img\_src string, url to the result image
thumbnail\_src string, url to a small-preview image
========================= =====================================================
.. table:: Parameter of the **videos** media type:
:width: 100%
========================= =====================================================
result-parameter information
------------------------- -----------------------------------------------------
template is set to ``videos.html``
========================= =====================================================
url string, url of the result
title string, title of the result
content *(not implemented yet)*
publishedDate :py:class:`datetime.datetime`, time of publish
thumbnail string, url to a small-preview image
========================= =====================================================
.. _magnetlink: https://en.wikipedia.org/wiki/Magnet_URI_scheme
.. table:: Parameter of the **torrent** media type:
:width: 100%
========================= =====================================================
result-parameter information
------------------------- -----------------------------------------------------
template is set to ``torrent.html``
========================= =====================================================
url string, url of the result
title string, title of the result
content string, general result-text
publishedDate :py:class:`datetime.datetime`,
time of publish *(not implemented yet)*
seed int, number of seeder
leech int, number of leecher
filesize int, size of file in bytes
files int, number of files
magnetlink string, magnetlink_ of the result
torrentfile string, torrentfile of the result
========================= =====================================================
.. table:: Parameter of the **map** media type:
:width: 100%
========================= =====================================================
result-parameter information
------------------------- -----------------------------------------------------
template is set to ``map.html``
========================= =====================================================
url string, url of the result
title string, title of the result
content string, general result-text
publishedDate :py:class:`datetime.datetime`, time of publish
latitude latitude of result (in decimal format)
longitude longitude of result (in decimal format)
boundingbox boundingbox of result (array of 4. values
``[lat-min, lat-max, lon-min, lon-max]``)
geojson geojson of result (https://geojson.org/)
osm.type type of osm-object (if OSM-Result)
osm.id id of osm-object (if OSM-Result)
address.name name of object
address.road street name of object
address.house_number house number of object
address.locality city, place of object
address.postcode postcode of object
address.country country of object
========================= =====================================================
A engine can append result-items of different media-types and different
result-types to the result list. The list of the result items is render to HTML
by templates. For more details read section :ref:`engine results`.

2
docs/dev/searxng_extra/update.rst
View file

@ -61,6 +61,8 @@ Scripts to update static data in :origin:`searx/data/`
:members:
.. _update_osm_keys_tags.py:
``update_osm_keys_tags.py``
===========================

466
docs/src/searx.results.rst Normal file
View file

@ -0,0 +1,466 @@
.. _engine results:
.. _searx.results:
==============
Engine Results
==============
.. automodule:: searx.results
:members:
The result items are organized in the :py:obj:`container.ResultContainer` and
rendered in the :ref:`result template macros` and :ref:`result template files`.
.. contents:: Contents
:depth: 2
:local:
:backlinks: entry
.. _standard result:
Result items
============
A result **item** is a python dictionary with dedicated keys and values. In the
result list a **standard result type** is identified by the existence of the key
``url``. Other **result types** are:
- :py:obj:`searx.results.suggestion`
- :py:obj:`searx.results.answer`
- :py:obj:`searx.results.correction`
- :py:obj:`searx.results.infobox`
The **standard result type**:
.. code:: python
results.append({
'template' : str,
# result_header
'url' : str,
'title' : str,
'content' : str,
'img_src' : str,
'thumbnail' : str,
# result_sub_header
'publishedDate' : datetime.datetime,
'length' : time.struct_time,
'author' : str,
'metadata' : str,
})
template : ``str``
:reF:`Media type <result media types>` of the result item. Name of the
:ref:`template file <result template files>` from :origin:`result_templates
<searx/templates/simple/result_templates>`. If unset, ``default.html`` is
used.
.. hint::
Each **standard result type** of an engine can be of different
:reF:`media-types <result media types>`.
.. _result template macros:
Result template macros
======================
.. _macro result_header:
``result_header``
-----------------
Execpt ``image.html`` this macro is used in all :ref:`result template files`.
Fields used in the template :origin:`macro result_header
<searx/templates/simple/macros.html>`:
url : ``str``
Link URL of the result item.
title : ``str``
Link title of the result item.
img_src, thumbnail : ``str``
URL of a image or thumbnail that is displayed in the result item.
.. _macro result_sub_header:
``result_sub_header``
---------------------
Execpt ``image.html`` this macro is used in all :ref:`result template files`.
Fields used in the template :origin:`macro result_sub_header
<searx/templates/simple/macros.html>`:
publishedDate : :py:obj:`datetime.datetime`
The date on which the object was published.
length: :py:obj:`time.struct_time`
Playing duration in seconds.
author : ``str``
Author of the title.
metadata : ``str``
Miscellaneous metadata.
.. _engine_data:
``engine_data_form``
--------------------
The ``engine_data_form`` macro is used in :origin:`results,html
<searx/templates/simple/results.html>` in a HTML ``<form/>`` element. The
intention of this macro is to pass data of a engine from one :py:obj:`response
<searx.engines.demo_online.response>` to the :py:obj:`searx.search.SearchQuery`
of the next :py:obj:`request <searx.engines.demo_online.request>`.
.. hint::
The engine-data values are transfered to the next request when the user press
the "next page" button. When a new search request is made, the enigine-data
are removed from the client request.
To pass data, engine's response handler can append result items of typ
``engine_data``. This is by example used to pass a token from the response to
the next request:
.. code:: python
def response(resp):
...
results.append({
'engine_data': token,
'key': 'next_page_token',
})
...
return results
def request(query, params):
page_token = params['engine_data'].get('next_page_token')
.. _result media types:
.. _result template files:
Result template files
=====================
The **media types** of the **standard result type** are the template files in
the :origin:`result_templates <searx/templates/simple/result_templates>`.
``default.html``
----------------
Displays result fields from:
- :ref:`macro result_header` and
- :ref:`macro result_sub_header`
Additional fields used in the :origin:`default.html
<searx/templates/simple/result_templates/default.html>`:
content : ``str``
General text of the result item.
iframe_src : ``str``
URL of an embedded ``<iframe>`` / the frame is collapsible.
audio_src : uri,
URL of an embedded ``<audio controls>``.
``code.html``
-------------
Displays result fields from:
- :ref:`macro result_header` and
- :ref:`macro result_sub_header`
Additional fields used in the :origin:`code.html
<searx/templates/simple/result_templates/code.html>`:
content : ``str``
Description of the code fragment.
codelines : ``[line1, line2, ...]``
Lines of the code fragment.
code_language : ``str``
Name of the code language, the value is passed to
:py:obj:`pygments.lexers.get_lexer_by_name`.
repository : ``str``
URL of the repository of the code fragment.
``images.html``
---------------
Fields used in the :origin:`images.html
<searx/templates/simple/result_templates/images.html>`:
title : ``str``
Title of the image.
thumbnail_src : ``str``
URL of a preview of the image.
img_src : ``str``
URL of the full size image.
Image labels
~~~~~~~~~~~~
content: ``str``
Description of the image.
author: ``str``
Name of the author of the image.
img_format : ``str``
Format of the image.
source : ``str``
Source of the image.
url : ``str``
URL of the page from where the images comes from (source).
``videos.html``
---------------
Displays result fields from:
- :ref:`macro result_header` and
- :ref:`macro result_sub_header`
Additional fields used in the :origin:`videos.html
<searx/templates/simple/result_templates/videos.html>`:
iframe_src : ``str``
URL of an embedded ``<iframe>`` / the frame is collapsible.
content : ``str``
Description of the code fragment.
``map.html``
------------
.. _GeoJSON: https://en.wikipedia.org/wiki/GeoJSON
.. _Leaflet: https://github.com/Leaflet/Leaflet
.. _bbox: https://wiki.openstreetmap.org/wiki/Bounding_Box
.. _HTMLElement.dataset: https://developer.mozilla.org/en-US/docs/Web/API/HTMLElement/dataset
.. _Nominatim: https://nominatim.org/release-docs/latest/
.. _Lookup: https://nominatim.org/release-docs/latest/api/Lookup/
.. _place_id is not a persistent id:
https://nominatim.org/release-docs/latest/api/Output/#place_id-is-not-a-persistent-id
.. _perma_id: https://wiki.openstreetmap.org/wiki/Permanent_ID
.. _country code: https://wiki.openstreetmap.org/wiki/Country_code
Displays result fields from:
- :ref:`macro result_header` and
- :ref:`macro result_sub_header`
Additional fields used in the :origin:`map.html
<searx/templates/simple/result_templates/map.html>`:
content : ``str``
Description of the item.
address_label : ``str``
Label of the address / default ``_('address')``.
geojson : GeoJSON_
Geometries mapped to HTMLElement.dataset_ (``data-map-geojson``) and used by
Leaflet_.
boundingbox : ``[ min-lon, min-lat, max-lon, max-lat]``
A bbox_ area defined by min longitude , min latitude , max longitude and max
latitude. The bounding box is mapped to HTMLElement.dataset_
(``data-map-boundingbox``) and is used by Leaflet_.
longitude, latitude : ``str``
Geographical coordinates, mapped to HTMLElement.dataset_ (``data-map-lon``,
``data-map-lat``) and is used by Leaflet_.
address : ``{...}``
A dicticonary with the address data:
.. code:: python
address = {
'name' : str, # name of object
'road' : str, # street name of object
'house_number' : str, # house number of object
'postcode' : str, # postcode of object
'country' : str, # country of object
'country_code' : str,
'locality' : str,
}
country_code : ``str``
`Country code`_ of the object.
locality : ``str``
The name of the city, town, township, village, borough, etc. in which this
object is located.
links : ``[link1, link2, ...]``
A list of links with labels:
.. code:: python
links.append({
'label' : str,
'url' : str,
'url_label' : str, # set by some engines but unused (oscar)
})
data : ``[data1, data2, ...]``
A list of additional data, shown in two columns and containing a label and
value.
.. code:: python
data.append({
'label' : str,
'value' : str,
'key' : str, # set by some engines but unused
})
type : ``str`` # set by some engines but unused (oscar)
Tag label from :ref:`OSM_KEYS_TAGS['tags'] <update_osm_keys_tags.py>`.
type_icon : ``str`` # set by some engines but unused (oscar)
Type's icon.
osm : ``{...}``
OSM-type and OSM-ID, can be used to Lookup_ OSM data (Nominatim_). There is
also a discussion about "`place_id is not a persistent id`_" and the
perma_id_.
.. code:: python
osm = {
'type': str,
'id': str,
}
type : ``str``
Type of osm-object (if OSM-Result).
id :
ID of osm-object (if OSM-Result).
.. hint::
The ``osm`` property is set by engine ``openstreetmap.py``, but it is not
used in the ``map.html`` template yet.
``products.html``
-----------------
Displays result fields from:
- :ref:`macro result_header` and
- :ref:`macro result_sub_header`
Additional fields used in the :origin:`products.html
<searx/templates/simple/result_templates/products.html>`:
content : ``str``
Description of the product.
price : ``str``
The price must include the currency.
shipping : ``str``
Shipping details.
source_country : ``str``
Place from which the shipment is made.
``torrent.html``
----------------
.. _magnet link: https://en.wikipedia.org/wiki/Magnet_URI_scheme
.. _torrent file: https://en.wikipedia.org/wiki/Torrent_file
Displays result fields from:
- :ref:`macro result_header` and
- :ref:`macro result_sub_header`
Additional fields used in the :origin:`torrent.html
<searx/templates/simple/result_templates/torrent.html>`:
magnetlink:
URL of the `magnet link`_.
torrentfile
URL of the `torrent file`_.
seed : ``int``
Number of seeders.
leech : ``int``
Number of leecher
filesize : ``int``
Size in Bytes (rendered to human readable unit of measurement).
files : ``int``
Number of files.
Suggestion results
==================
.. automodule:: searx.results.suggestion
:members:
Answer results
==============
.. automodule:: searx.results.answer
:members:
Correction results
==================
.. automodule:: searx.results.correction
:members:
Infobox results
===============
.. automodule:: searx.results.infobox
:members:
Result container
================
.. automodule:: searx.results.container
:members:
results.core
============
.. automodule:: searx.results.core
:members:

3
searx/results/__init__.py Normal file
View file

@ -0,0 +1,3 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Implementation of the result container and the result types."""

30
searx/results/answer.py Normal file
View file

@ -0,0 +1,30 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Answer item in the result list. The answer result item is used in
the :origin:`results.html <searx/templates/simple/results.html>` template.
A answer item is a dictionary type with dedicated keys and values. In the
result list a answer item is identified by the existence of the key
``suggestion``.
.. code:: python
results.append({
'answer' : str,
'url' : str,
})
answer : ``str``
The answer string append by the engine.
url : ``str``
A link that is related to the answer (e.g. the origin of the answer)
"""
class Answers(dict):
"""Dictionary of answers in the :py:obj:`.container.ResultContainer`"""
def add(self, result):
self[result['answer']] = result

230
searx/results.py → searx/results/container.py
View file

@ -1,158 +1,95 @@
import re
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""ResultContainer
"""
from collections import defaultdict
from operator import itemgetter
from threading import RLock
from typing import List, NamedTuple, Set
from urllib.parse import urlparse, unquote
from typing import List, Set
from urllib.parse import urlparse
from searx import logger
from searx.engines import engines
from searx.metrics import histogram_observe, counter_add, count_error
from .core import (
WHITESPACE_REGEX,
Timing,
UnresponsiveEngine,
result_content_len,
result_score,
compare_urls,
)
CONTENT_LEN_IGNORED_CHARS_REGEX = re.compile(r'[,;:!?\./\\\\ ()-_]', re.M | re.U)
WHITESPACE_REGEX = re.compile('( |\t|\n)+', re.M | re.U)
from .infobox import Infoboxes, merge_two_infoboxes
from .suggestion import Suggestions
from .answer import Answers
from .correction import Corrections
# return the meaningful length of the content for a result
def result_content_len(content):
if isinstance(content, str):
return len(CONTENT_LEN_IGNORED_CHARS_REGEX.sub('', content))
else:
return 0
def is_suggestion(result):
"""Returns ``True`` if result type is :py:obj:`.suggestion`, otherwise
``False``"""
return 'suggestion' in result
def compare_urls(url_a, url_b):
"""Lazy compare between two URL.
"www.example.com" and "example.com" are equals.
"www.example.com/path/" and "www.example.com/path" are equals.
"https://www.example.com/" and "http://www.example.com/" are equals.
Args:
url_a (ParseResult): first URL
url_b (ParseResult): second URL
Returns:
bool: True if url_a and url_b are equals
"""
# ignore www. in comparison
if url_a.netloc.startswith('www.'):
host_a = url_a.netloc.replace('www.', '', 1)
else:
host_a = url_a.netloc
if url_b.netloc.startswith('www.'):
host_b = url_b.netloc.replace('www.', '', 1)
else:
host_b = url_b.netloc
if host_a != host_b or url_a.query != url_b.query or url_a.fragment != url_b.fragment:
return False
# remove / from the end of the url if required
path_a = url_a.path[:-1] if url_a.path.endswith('/') else url_a.path
path_b = url_b.path[:-1] if url_b.path.endswith('/') else url_b.path
return unquote(path_a) == unquote(path_b)
def is_answer(result):
"""Returns ``True`` if result type is :py:obj:`.answer`, otherwise ``False``"""
return 'answer' in result
def merge_two_infoboxes(infobox1, infobox2):
# get engines weights
if hasattr(engines[infobox1['engine']], 'weight'):
weight1 = engines[infobox1['engine']].weight
else:
weight1 = 1
if hasattr(engines[infobox2['engine']], 'weight'):
weight2 = engines[infobox2['engine']].weight
else:
weight2 = 1
if weight2 > weight1:
infobox1['engine'] = infobox2['engine']
infobox1['engines'] |= infobox2['engines']
if 'urls' in infobox2:
urls1 = infobox1.get('urls', None)
if urls1 is None:
urls1 = []
for url2 in infobox2.get('urls', []):
unique_url = True
parsed_url2 = urlparse(url2.get('url', ''))
entity_url2 = url2.get('entity')
for url1 in urls1:
if (entity_url2 is not None and url1.get('entity') == entity_url2) or compare_urls(
urlparse(url1.get('url', '')), parsed_url2
):
unique_url = False
break
if unique_url:
urls1.append(url2)
infobox1['urls'] = urls1
if 'img_src' in infobox2:
img1 = infobox1.get('img_src', None)
img2 = infobox2.get('img_src')
if img1 is None:
infobox1['img_src'] = img2
elif weight2 > weight1:
infobox1['img_src'] = img2
if 'attributes' in infobox2:
attributes1 = infobox1.get('attributes')
if attributes1 is None:
infobox1['attributes'] = attributes1 = []
attributeSet = set()
for attribute in attributes1:
label = attribute.get('label')
if label not in attributeSet:
attributeSet.add(label)
entity = attribute.get('entity')
if entity not in attributeSet:
attributeSet.add(entity)
for attribute in infobox2.get('attributes', []):
if attribute.get('label') not in attributeSet and attribute.get('entity') not in attributeSet:
attributes1.append(attribute)
if 'content' in infobox2:
content1 = infobox1.get('content', None)
content2 = infobox2.get('content', '')
if content1 is not None:
if result_content_len(content2) > result_content_len(content1):
infobox1['content'] = content2
else:
infobox1['content'] = content2
def is_correction(result):
"""Returns ``True`` if result type is :py:obj:`.correction`, otherwise
``False``"""
return 'correction' in result
def result_score(result):
weight = 1.0
for result_engine in result['engines']:
if hasattr(engines[result_engine], 'weight'):
weight *= float(engines[result_engine].weight)
occurences = len(result['positions'])
return sum((occurences * weight) / position for position in result['positions'])
def is_infobox(result):
"""Returns ``True`` if result type is :py:obj:`.infobox`, otherwise ``False``"""
return 'infobox' in result
class Timing(NamedTuple):
engine: str
total: float
load: float
def is_number_of_results(result):
"""Returns ``True`` if result type is ``number_of_results``, otherwise
``False``"""
return 'number_of_results' in result
class UnresponsiveEngine(NamedTuple):
engine: str
error_type: str
suspended: bool
def is_engine_data(result):
"""Returns ``True`` if result type is :ref:`engine_data`, otherwise ``False``"""
return 'engine_data' in result
def is_standard_result(result):
"""Returns ``True`` if result type is a :ref:`standard result <standard
result>`, otherwise ``False``"""
return 'url' in result
class ResultContainer:
"""docstring for ResultContainer"""
"""A container to organize the result items and the various result types. New
results can be added by :py:obj:`ResultContainer.extend`.
To be clear, a result-type is not a special python data-type, a result is
always a python dicticonary. The result-type is determined by the presence
of one of the following keys (in that order, first match wins)
1. suggestion: :py:obj:`.suggestion`
2. answer: :py:obj:`.answer`
3. correction: :py:obj:`.correction`
4. infobox: :py:obj:`.infobox`
5. number_of_results: Number of results origin engine has.
.. code:: python
results.append({
'number_of_results' : int,
})
6. engine_data: used to pass :ref:`engine_data <engine_data>` to next request.
7. url: :ref:`standard result <standard result>`
"""
__slots__ = (
'_merged_results',
@ -174,10 +111,10 @@ class ResultContainer:
def __init__(self):
super().__init__()
self._merged_results = []
self.infoboxes = []
self.suggestions = set()
self.answers = {}
self.corrections = set()
self.infoboxes = Infoboxes()
self.suggestions = Suggestions()
self.answers = Answers()
self.corrections = Corrections()
self._number_of_results = []
self.engine_data = defaultdict(dict)
self._closed = False
@ -188,7 +125,8 @@ class ResultContainer:
self.on_result = lambda _: True
self._lock = RLock()
def extend(self, engine_name, results):
def extend(self, engine_name, results): # pylint: disable=too-many-branches
"""Add a result item to the container."""
if self._closed:
return
@ -196,19 +134,19 @@ class ResultContainer:
error_msgs = set()
for result in list(results):
result['engine'] = engine_name
if 'suggestion' in result and self.on_result(result):
if is_suggestion(result) and self.on_result(result):
self.suggestions.add(result['suggestion'])
elif 'answer' in result and self.on_result(result):
self.answers[result['answer']] = result
elif 'correction' in result and self.on_result(result):
elif is_answer(result) and self.on_result(result):
self.answers.add(result)
elif is_correction(result) and self.on_result(result):
self.corrections.add(result['correction'])
elif 'infobox' in result and self.on_result(result):
elif is_infobox(result) and self.on_result(result):
self._merge_infobox(result)
elif 'number_of_results' in result and self.on_result(result):
elif is_number_of_results(result) and self.on_result(result):
self._number_of_results.append(result['number_of_results'])
elif 'engine_data' in result and self.on_result(result):
elif is_engine_data(result) and self.on_result(result):
self.engine_data[engine_name][result['key']] = result['engine_data']
elif 'url' in result:
elif is_standard_result(result):
# standard result (url, title, content)
if not self._is_valid_url_result(result, error_msgs):
continue
@ -313,7 +251,6 @@ class ResultContainer:
if result_template != 'images.html':
# not an image, same template, same url : it's a duplicate
return merged_result
else:
# it's an image
# it's a duplicate if the parsed_url, template and img_src are differents
if result.get('img_src', '') == merged_result.get('img_src', ''):
@ -363,6 +300,7 @@ class ResultContainer:
categoryPositions = {}
for res in results:
# pylint: disable=fixme
# FIXME : handle more than one category per engine
engine = engines[res['engine']]
res['category'] = engine.categories[0] if len(engine.categories) > 0 else ''
@ -389,7 +327,7 @@ class ResultContainer:
# update every index after the current one
# (including the current one)
for k in categoryPositions:
for k in categoryPositions: # pylint: disable=consider-using-dict-items
v = categoryPositions[k]['index']
if v >= index:
categoryPositions[k]['index'] = v + 1

80
searx/results/core.py Normal file
View file

@ -0,0 +1,80 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Core methods
"""
# pylint: disable=too-few-public-methods
import re
from urllib.parse import unquote
from typing import NamedTuple
from searx.engines import engines
CONTENT_LEN_IGNORED_CHARS_REGEX = re.compile(r'[,;:!?\./\\\\ ()-_]', re.M | re.U)
WHITESPACE_REGEX = re.compile('( |\t|\n)+', re.M | re.U)
class Timing(NamedTuple): # pylint: disable=missing-class-docstring
engine: str
total: float
load: float
class UnresponsiveEngine(NamedTuple): # pylint: disable=missing-class-docstring
engine: str
error_type: str
suspended: bool
def result_content_len(content):
"""Return the meaningful length of the content for a result."""
if isinstance(content, str):
return len(CONTENT_LEN_IGNORED_CHARS_REGEX.sub('', content))
return 0
def result_score(result):
weight = 1.0
for result_engine in result['engines']:
if hasattr(engines[result_engine], 'weight'):
weight *= float(engines[result_engine].weight)
occurences = len(result['positions'])
return sum((occurences * weight) / position for position in result['positions'])
def compare_urls(url_a, url_b):
"""Lazy compare between two URL.
"www.example.com" and "example.com" are equals.
"www.example.com/path/" and "www.example.com/path" are equals.
"https://www.example.com/" and "http://www.example.com/" are equals.
Args:
url_a (ParseResult): first URL
url_b (ParseResult): second URL
Returns:
bool: True if url_a and url_b are equals
"""
# ignore www. in comparison
if url_a.netloc.startswith('www.'):
host_a = url_a.netloc.replace('www.', '', 1)
else:
host_a = url_a.netloc
if url_b.netloc.startswith('www.'):
host_b = url_b.netloc.replace('www.', '', 1)
else:
host_b = url_b.netloc
if host_a != host_b or url_a.query != url_b.query or url_a.fragment != url_b.fragment:
return False
# remove / from the end of the url if required
path_a = url_a.path[:-1] if url_a.path.endswith('/') else url_a.path
path_b = url_b.path[:-1] if url_b.path.endswith('/') else url_b.path
return unquote(path_a) == unquote(path_b)

39
searx/results/correction.py Normal file
View file

@ -0,0 +1,39 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Correction item in the result list. The correction result item is used in
the :origin:`results.html <searx/templates/simple/results.html>` template.
A correction item is a dictionary type with dedicated keys and values. In the
result list a answer item is identified by the existence of the key
``correction``.
.. code:: python
results.append({
'correction' : str,
})
The context ``corrections`` of the HTML template is a set of dictionaries:
.. code:: python
corrections = [
{
'url' : str,
'title' : str,
},
{...},
...
]
url : ``str``
The search URL for the correction
title : ``str``
The 'correction' string append by the engine.
"""
class Corrections(set):
"""Set of corrections in the :py:obj:`.container.ResultContainer`"""

157
searx/results/infobox.py Normal file
View file

@ -0,0 +1,157 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Infobox item in the result list. The infobox result item is used in the
:origin:`infobox.html <searx/templates/simple/infobox.html>` template.
A infobox item is a dictionary type with dedicated keys and values. In the
result list a infobox item is identified by the existence of the key ``infobox``.
.. code:: python
results.append({
'infobox' : str,
'id' : str,
'content' : str,
'img_src' : str,
'urls' : [url, ...],
'attributes' : [attribute, ...],
'relatedTopics' : [topic, ...],
'engine' : engine,
})
infobox : ``str``
Name of the infobox (mandatory).
id : ``str``
URL of the infobox. Will be used to merge infoboxes.
content : ``str``
Content of the infobox (the description)
img_src:
URL of the image to show in the infobox
urls : ``[url, ...]``
A list of dictionaries with links shown in the infobox. A **url** item in the
``infobox.urls`` list is a dicticonary:
.. code:: python
url = {
'title' : str,
'url' : str,
'entity' : str, # set by some engines but unused
'official' : bool, # set by some engines but unused (oscar)
}
attributes : ``[attribute, ...]``
A **attribute** item in the ``infobox.attributes`` list is a dictionary:
.. code:: python
attribute = {
'label' : str,
'value' : str,
'image' : {
'src': str,
'alt': str,
},
'entity' : str, # set by some engines but unused
}
relatedTopics : ``[topic, ...]``
A **topic** item in the ``infobox.relatedTopics`` list is a dictionary:
.. code:: python
topic = {
'suggestion' : str,
'name' : str, # set by some engines but unused
}
"""
from urllib.parse import urlparse
from searx.engines import engines
from .core import (
result_content_len,
compare_urls,
)
class Infoboxes(list):
"""List of infobox items in the :py:obj:`.container.ResultContainer`"""
def merge_two_infoboxes(infobox1, infobox2):
# pylint: disable=too-many-branches, too-many-statements
# get engines weights
if hasattr(engines[infobox1['engine']], 'weight'):
weight1 = engines[infobox1['engine']].weight
else:
weight1 = 1
if hasattr(engines[infobox2['engine']], 'weight'):
weight2 = engines[infobox2['engine']].weight
else:
weight2 = 1
if weight2 > weight1:
infobox1['engine'] = infobox2['engine']
infobox1['engines'] |= infobox2['engines']
if 'urls' in infobox2:
urls1 = infobox1.get('urls', None)
if urls1 is None:
urls1 = []
for url2 in infobox2.get('urls', []):
unique_url = True
parsed_url2 = urlparse(url2.get('url', ''))
entity_url2 = url2.get('entity')
for url1 in urls1:
if (entity_url2 is not None and url1.get('entity') == entity_url2) or compare_urls(
urlparse(url1.get('url', '')), parsed_url2
):
unique_url = False
break
if unique_url:
urls1.append(url2)
infobox1['urls'] = urls1
if 'img_src' in infobox2:
img1 = infobox1.get('img_src', None)
img2 = infobox2.get('img_src')
if img1 is None:
infobox1['img_src'] = img2
elif weight2 > weight1:
infobox1['img_src'] = img2
if 'attributes' in infobox2:
attributes1 = infobox1.get('attributes')
if attributes1 is None:
infobox1['attributes'] = attributes1 = []
attributeSet = set()
for attribute in attributes1:
label = attribute.get('label')
if label not in attributeSet:
attributeSet.add(label)
entity = attribute.get('entity')
if entity not in attributeSet:
attributeSet.add(entity)
for attribute in infobox2.get('attributes', []):
if attribute.get('label') not in attributeSet and attribute.get('entity') not in attributeSet:
attributes1.append(attribute)
if 'content' in infobox2:
content1 = infobox1.get('content', None)
content2 = infobox2.get('content', '')
if content1 is not None:
if result_content_len(content2) > result_content_len(content1):
infobox1['content'] = content2
else:
infobox1['content'] = content2

41
searx/results/suggestion.py Normal file
View file

@ -0,0 +1,41 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Suggestion item in the result list. The suggestion result item is used in
the :origin:`infobox.html <searx/templates/simple/results.html>` template.
A sugestion item is a dictionary type with dedicated keys and values. In the
result list a suggestion item is identified by the existence of the key
``suggestion``.
.. code:: python
results.append({
'suggestion' : str,
})
The context ``suggestions`` of the HTML template is a set of dictionaries:
.. code:: python
suggestions = [
{
'url' : str,
'title' : str,
},
{...},
...
]
url : ``str``
The search URL for the suggestion
title : ``str``
The 'suggestion' string append by the engine.
"""
from typing import Set
class Suggestions(Set):
"""Set of suggestions in the :py:obj:`.container.ResultContainer`"""

2
searx/search/__init__.py
View file

@ -11,7 +11,7 @@ import flask
from searx import settings
from searx.answerers import ask
from searx.external_bang import get_bang_url
from searx.results import ResultContainer
from searx.results.container import ResultContainer
from searx import logger
from searx.plugins import plugins
from searx.search.models import EngineRef, SearchQuery

2
searx/search/checker/impl.py
View file

@ -16,7 +16,7 @@ import httpx
from searx import network, logger
from searx.utils import gen_useragent
from searx.results import ResultContainer
from searx.results.container import ResultContainer
from searx.search.models import SearchQuery, EngineRef
from searx.search.processors import EngineProcessor
from searx.metrics import counter_inc

2
searx/webapp.py
View file

@ -58,7 +58,7 @@ from searx import (
from searx import infopage
from searx.data import ENGINE_DESCRIPTIONS
from searx.results import Timing, UnresponsiveEngine
from searx.results.core import Timing, UnresponsiveEngine
from searx.settings_defaults import OUTPUT_FORMATS
from searx.settings_loader import get_default_settings_path
from searx.exceptions import SearxParameterException

2
tests/unit/test_results.py
View file

@ -1,6 +1,6 @@
# -*- coding: utf-8 -*-
from searx.results import ResultContainer
from searx.results.container import ResultContainer
from tests import SearxTestCase

2
tests/unit/test_webapp.py
View file

@ -3,7 +3,7 @@
import json
from urllib.parse import ParseResult
from mock import Mock
from searx.results import Timing
from searx.results.core import Timing
import searx.search.processors
from searx.search import Search